A decade after the end of apartheid, the well-being of South African children is still in a precarious state. Nearly 70% of the nation's Black African children live in households with incomes less than $2000 per year. The HIV/AIDS prevalence rate for pregnant women attending public antenatal services is over 30%. And, by 2010, 19% of South African children will have experienced the death of one or both parents, half due to AIDS. In response to this crisis, the overarching objective of the proposed project is to conduct a short-term, longitudinal, multi-level study of 6000 7- to 10-year-olds and their parents/parent surrogates in 60 urban and rural South African communities in KwaZulu-Natal. We posit three specific aims. AIM 1. Examine the associations between a) a set of major household risk factors and a set of adverse childhood experiences; and between b) the occurrence and nature of adverse childhood experiences and child psychosocial, health and educational outcomes. AIM 2. Explore the degree to which selected factors at multiple levels moderate the influence of major household risk factors on adverse childhood experiences, and adverse childhood experiences on key child outcomes. AIM 3. Test the effects of a major social policy innovation Conditional Cash Transfers on household and childhood risk factors (directly) and children's well-being (indirectly). This project is a collaboration of researchers at New York University in the U.S. and the Human Sciences Research Council in South Africa in cooperation with the South African government and the World Bank. [unreadable] [unreadable] PUBLIC HEALTH RELEVANCE The results of this study have two main implications for public health. First, one set of results will indicate whether Conditional Cash Transfers can improve the health, education and well-being of poor South African children in high HIV/AIDS prevalence communities. Second, another set of results can inform the design of new public health and social policy strategies to support households in AIDS-affected communities. [unreadable] [unreadable] [unreadable]